1 Uncorrected Preprint Final Publication Details
Total Page:16
File Type:pdf, Size:1020Kb
Uncorrected Preprint Final publication details: https://doi.org/10.1016/j.jasrep.2020.102477 Received 4 June 2020; Accepted 7 July 2020; 2352-409/© 2020. THE TIP CROSS-SECTIONAL AREAS OF POISONED BONE ARROWHEADS FROM SOUTHERN AFRICA Marlize Lombard Palaeo-Research Institute, University of Johannesburg, P.O. Box 524, Auckland Park, ZA-2006, South Africa. E-mail: [email protected] Abstract Current hunter-gatherers from the Kalahari in southern Africa are well-known for their use of poisoned arrowheads, and it is assumed that this tradition spanned most of the Holocene in the region. Recent archaeological work, however, indicates that the techno-behaviour may have originated sometime during the Pleistocene. Tracing the use of poisoned arrowheads through time is not an easy task. Here I explore the use of the tip cross-sectional area (TSCA) metric to analyse relatively large samples of bone points that are ethno-historically associated with Kalahari San poisoned arrow hunting. I add the southern African poisoned bone arrowhead TSCA range to the previous ranges established for North American atlatl dart tips, North American arrowheads and large thrusting spears. Based on the results obtained from 445 artefacts spanning historical, Later and Middle Stone Age phases, I show that poisoned bone arrowheads may have been in use in southern Africa throughout the last 60 000 years, and that a methodical effort to trace stone-tipped poison arrowheads may be warranted. Keywords: poisoned arrows, bow hunting, Kalahari San, Middle Stone Age, arrow morphology Introduction Today, the Kalahari San of southern Africa are renowned for their use of relatively small bows and light-weight poisoned arrows, and there is a plethora of ethno-historical literature on the topic (Sparrman 1785; Stow 1905; Theal 1919; Schapera 1927; Dunn 1931; Lebzelter 1934; Vinnicombe 1971; Shaw et al. 1963; Marshall 1976; Lee 1979; Wiessner 1983; Deacon 1992; Webley 1994; Nadler 2005; Wadley et al. 2015; Chaboo et al. 2016; Backwell et al. 2018; Hitchcock et al. 2019). The origin of this hunting system, is however, a matter of debate (d’Errico et al. 2012a,b; Evans 2012; Bradfield et al. 2015). The tip cross-sectional area (TCSA) metric is a ballistically significant standard 1 that works well for discriminating between, North American atlatl or spear-thrower dart tips, North American arrowheads and large experimental stabbing spears (Shea 2006; Sisk & Shea 2010). Using this metric, Shea (2006) concluded that his results did not support the hypothesis of the widespread use of stone-tipped atlatl darts or arrows in Africa before ~40 ka – an interpretation supported by Villa and Lenoir (2006). Bone points, although plausible weapon tips, were excluded from his analysis, because at the time they were considered to be “too few and too rare to be the focus of an investigation into broader patterns of projectile point origins and variability” (Shea 2006: 824). By now, the TCSA metric has been used widely for interpreting variation in stone-tipped weapon- delivery systems and in experimental replications of such systems (see several chapters in Iovita & Sano 2016; also see discussion in Villa & Lenoir 2006). The metric is useful, because it represents the part of the weapon that cuts the animal’s hide, creating a hole for the shaft to enter (Hughes 1998), and it may influence weapon flight and penetration dynamics (Maki 2013). Regardless of point type, only maximum width and thickness measurements are needed for calculating the TCSA metric (0.5 x maximum width x maximum thickness = TCSA value). This allows for the largest possible sample sizes to be compared directly with each other, and for studies to be easily replicated. Sisk and Shea (2011) used TCSA values in combination with the tip cross-sectional perimeter (TCSP), which they deemed a more accurate proxy of the force needed to penetrate a target to a lethal depth. The drawback of the TCSP approach (when it comes to the interpretation of archaeological material as opposed to modern experimentation), is that penetration depth is invariably also affected by the mass of the shaft and the traits of the propelling mechanism, which cannot be known for archaeological bone- or stone-tipped weapons. Also, when poison is used, depth of penetration does not have to be lethal. Instead, all that is needed, is for the hide to be cut so that the poison may enter the prey animal’s bloodstream – making the TCSA metric the most relevant. The TCSP metric is furthermore limited to stone points, and cannot be applied to backed geometrics (Sisk & Shea 2010), which currently represent the earliest evidence for bow hunting with stone-tipped or barbed arrows (Lombard & Phillipson 2010; Lombard 2011). TCSA analysis on the other hand, can be applied to these backed artefacts (Wadley & Mohapi 2008; Sano et al. 2019), as well as to bone points (Bradfield 2010). Brooks and colleagues (2006) emphasised weight as the most important aspect of weapon tips, based on the prediction that lighter projectiles require greater acceleration to achieve adequate penetration. Yet, similar to the TCSP metric, the weight of a weapon tip represents only a portion of a weapon’s mass, and the weight of the shaft cannot be known for archaeological material where these elements do not preserve. I also need to draw attention to the fact that any metric approach to weapon function, whether TCSA, TCSP, penetrating angle or mass, is only able to assess the potential of artefacts to function in the 2 context of different weapon-delivery systems. These approaches can never demonstrate such use empirically. For example, Ötzi’s well-known stone-bladed knife seen on its own, would be interpreted as a spear tip with a TCSA value of 92 (measured from the drawing of Wierer et al. 2018: 6). Yet, because it was found hafted, we know it was a knife blade instead. Two more stone points found with Ötzi have TCSA values of 40 and 45 respectively, which would fall in the lower range of North American atlatls and the higher range of North American arrows. But, because they were found hafted in arrow shafts notched to fit a bowstring, we know their explicit function. For direct evidence that any ancient, haftless stone artefact functioned as a weapon tip, detailed and multi-stranded use-trace analyses are needed (Lombard 2005, 2006, 2011; Lombard & Phillipson 2010; also see Newman & Moore 2013; Clarkson 2016; Hutchings 2016). The overlap in Shea’s (2006) dart and arrow tip categories also reveals the false sense of security in purely quantitative studies. When dealing with the human past, numbers alone can seldom reveal the nuances necessary for a deep understanding of techno-behaviours – for that a measure of qualitative assessment and interpretation is required. That being said, TCSA analysis is a pragmatic approach for analysing considerable sample sizes across spatiotemporal boundaries in a directly comparable manner, and therefore apt for building broad hypotheses about ancient weapon-delivery systems (Sano et al. 2016). This is especially true when interpretations of the results are backed up by ethno-historical, experimental and use-trace data. Here I argue that it may also be useful for assessing a tool class’s potential to function as a poisoned arrowhead, and that the TCSA range for this category of weapon tips has yet to be established. In the southern African context, the most secure starting point for such investigation is bone points. Kalahari San hunter-gatherers still use bone, as well as very small iron tips, for poisoned arrow hunting (Wadley et al. 2015; Alam 2019). Many museums across the world now have relatively large collections of San bow-hunting kits with quivers full of ethno-historically collected poisoned bone- tipped arrows (Fig. 1). The function of the bone points themselves is therefore incontestable. Although arrow link-shafts, needles and awls cannot be excluded from the bone point category, careful morphological and use-trace work shows that in most cases, relatively thin and straight Later Stone Age bone points served as arrowheads (Bradfield 2012, 2015, 2016). What is more, several archaeological pieces also still have visible poison remains. 3 Figure 1. 1a: Arrows collected from the Cape of Good Hope, South Africa, by Atherstone in 1863 now housed in the Kew Economic Botany Collection in London, UK. 1b: Bone point arrowheads that became separated from the Atherstone arrows and poison collected with them. 1c: The hunter-gatherer quiver and arrows that were collected by Sparrman in 1777 from San hunter-gatherers in the Eastern Cape, South Africa, now curated at the Ethnographic Museum in Stockholm, Sweden. 1d & e: Details of other ethno-historically collected poisoned bone arrowheads from southern Africa housed at the Ethnographic Museum in Stockholm. (All photographs by Anders Högberg©.) Initially it was thought that poisoned bone arrowheads were used in southern Africa only after ~8 ka (ka=thousand years old/ago) (Inskeep 1987; Noli 1993). The last decade, however, saw some indication that archaeological bone points from the region, dating to more than 30 ka, may have been used to tip poisoned arrows (d’Errico et al. 2012; Robbins et al. 2012). Most recently, we reported that a bone point of >60 ka, from Klasies River Mouth in the Eastern Cape, South Africa, has a residue that may be poison based on its distribution and the interpretation of preliminary non-destructive Raman spectra (Bradfield et al. 2020). This point, as well as one from Sibudu Cave in KwaZulu- Natal, South Africa, of similar age (Bradfield & Lombard 2011; Backwell et al. 2018), also displays macro- and micro-fractures that are consistent with stresses caused by high-velocity impact, such as arrow use (Bradfield et al. 2020). With this contribution I aim to establish the TCSA range for poisoned bone arrowheads from southern Africa to complement the set of standards currently in use as established by Shea (2006).